This week, sections of the print media barely passed the test on the journalism of attribution, the craft’s principal flag of authenticity.

This paper, for example, swung from pretty thin attributions to great citations. The Tuesday lead story, "Summit grills Raila over ‘topple’ remark," for example, was a must-read. But halfway through the page, a critical reader’s mind still asked, "Who’s doing the talking?" Yet, Thursday’s lead story, "Raila’s 150,000 homes plan killed," scooped fairly good marks on attribution, even if the sources remained anonymous.

What is this all about? This is about a reporter’s job, which is much like that of a midwife. A reporter’s job includes gathering information from a variety of sources. When he presents this information to his audience, the reporter is not the expert; his sources are. Which is why quotes or disputable facts must be attributed to sources.

Agreed, when a reporter storms into the newsroom from a beat with piles of notes, it is quite easy for him to get into a daze. Does he just vomit the entire story with no particular sequence, or does he only tell the reader the highlights of his story?

Yes, a reporter may have it at the back of his mind that he is only the conveyer belt, not the luggage; the messenger, not the message. But his effort in attributing his story is even further compounded by the fact that there is no clear-cut way of going about the business. He has, like, six options.

Much like a mechanic with his choice of tools. Most mechanics can do repair jobs with just one screwdriver, but sometimes they need a small star screwdriver; other times they need a flat-blade. Sometimes they meet a screw that requires one of those huge hex screwdrivers, they may as well use a hammer.

Such is the frequent dilemma of a reporter faced with attributing his story. There are times the writer needs to stick his attribution upfront. In the failed-housing-project story on Thursday, for example, this paper wrote, "A former senior Cabinet minister, who drew the East African Standard’s attention to the issue said: ‘Apart from petroleum and gas, this would have been the single biggest American private sector investment in the whole of the African continent.’’’ Let’s call this attribution Type A.

Then, there are times a writer rambles on till the end before he tells his reader who has said this. The story above had this, for example: "But in an endeavour to get a tax waiver, the investor was incessantly and unnecessarily frustrated by senior officials at the Treasury through meaningless delays," [a source in the US embassy] revealed. Let’s call this Type B.

There is still a third type, where a writer finds a logical place somewhere in the middle of the sentence to pause and put the attribution. "Yesterday, Lands minister Amos Kamunya said the Government was anxious to establish if the Skyview complaint should be viewed as "total fraud and fictitious,’’ the Nation wrote in its Thursday lead story, "AG on the spot over billions lost in court." Let’s call this Type C.

For each of these attributions, there are still two flavours – the direct and the indirect quote. The point is that a reporter has a variety of tools to employ in attributing his facts. Just how much attribution is necessary may be a matter of feel and house style.

The undisputed requirement, however, is that the reader needs to know who is doing the speaking.

But a skim through local stories reveals a hole in this crucial journalistic requirement.

The Tuesday story on Raila’s "topple" remark, for example, ran for 14 paragraphs without a single attribution one could put a finger on. A similar scenario was replayed in the Nation’s story, "AG on the spot over billions lost in court." The story kicked off with four potentially disputable paragraphs without a single attribution.

Another story in the same issue, "City Hall row over move to topple mayor," ran from start to finish with the only attribution given to an anonymous source. Which brings us to the razor-thin leeway journalism permits with regard to anonymous sources.

The journalism craft is riddled with "It’s a creation of the media" allegations. This is the reason the discipline of the craft requires that reporters name their sources. Yet, there are lots of instances when this may not be possible or prudent. Politicians, for example, float "trial balloons" and insist on anonymity. Reporters are advised to decline a story on that basis. A whistle blower might be afraid of losing a job, and may insist on anonymity as a condition for sharing information. Or a reporter may be writing a story on something sensitive, such as a person’s sexual orientation, and may want to protect his source from public humiliation.

The practice this week, however, revealed too many instances where reporters wrote contentious issues and slapped it all on "sources said". Such as the Nation’s City Hall story on Thursday, and the Standard’s Raila story on Tuesday.

Joe Lelyveld, executive editor of the New York Times, required that his reporters and editors ask themselves two questions before using anonymous sources. One: How much direct knowledge does the anonymous source have of the event? Two: What, if any, motive might the source have for misleading us, gilding the lily, or hiding important facts that might alter our impression of the information?

Only after reporters are satisfied by answers to these questions should they use an anonymous source. And, then, they must indicate to the audience how the source was in a position to know the facts. The rationale is simple: if journalists are truth seekers, they must appear to be honest and truthful with their audience.

Kodi Barth teaches journalism at United States International University-Nairobi

Copyright © 2004 . The Standard Ltd